Harvard researchers Daniel Gilbert, Gary King, and Stephen Pettigrew, along with University of Virginia professor Timothy Wilson, calculated that the number of failed replications in the OSC study was no greater than would be expected by chance, even if all 100 original findings were true.
Notes on verification
The claim accurately represents the published GKPW 2016 technical comment and is corroborated by multiple independent sources including the original Science publication. However, the statistical methodology and conclusion were disputed by the original OSC authors and other statisticians (e.g., Lakens), making this a contested rather than settled scientific claim. [tier=gold indep_score=0.983 clusters=3 claim_tier=notable]
Sources
- Study that undercut psych research got it wrong (seed:psychology_and_neuroscience)
- https://www.science.org/doi/10.1126/science.aad7243 (corroboration)
- http://daniellakens.blogspot.com/2016/03/the-statistical-conclusions-in-gilbert.html (corroboration)
- https://www.nature.com/articles/nature.2016.19503 (corroboration)